1. Field of the Invention
The present invention pertains to automatic gain control in a data-packet-based communication receiver, such as a receiver in a wireless CDMA 1xEVDO system or a receiver in a wireless 802.11 network.
2. Description of the Related Art
FIG. 1 illustrates a block diagram of a conventional automatic gain control (AGC) system 10. AGC is a signal-processing technique that is used, inter alia, by a communication receiver to dynamically compensate for widely varying channel gains encountered in various wireless and wire-line transmission channels 12. In a conventional approach, the AGC block 10 forms a loop by estimating 16 the signal strength of the received signal at the output of a variable-gain block 14, filtering 18 the estimate to smooth out the instantaneous variations (e.g., due to noise), comparing the results to a specified target value 19, and then changing the gain in variable-gain block 14 in a feedback fashion in an attempt to maintain the received signal strength at the specified target value 19. The gain-adjusted signal is then subsequently processed 20 to decode the information embedded in the received signal.
The loop filter 18 is usually designed to have a fixed time-constant that is chosen to achieve a balance between the contradictory requirements of broader loop bandwidth for faster AGC loop tracking of the input signal strength variations (due to changes in the gain in transmission channel 12) and narrower bandwidth to reduce noise. Any difference between the filtered estimate and the specified target value is assumed to have been caused by a change in gain in the transmission channel 12 from the last measurement. Accordingly, this difference is used to control the gain in block 14, in a feedback fashion. Under appropriate circumstances, this feedback process ensures stable tracking and compensation for changes in the channel gain in an iterative manner.
However, the present inventors have discovered that the fixed bandwidth of the conventional receiver AGC loop filter 18 typically is optimized for a received signal that is continuous in nature and does not work very well with a received signal that is discontinuous in nature (e.g., packet transmissions) or has step changes in its strength. This is because of the fixed time-constant of the AGC loop-filter, which is typically selected to keep the loop bandwidth narrow so as to keep the noise in the signal strength estimates 16 from causing spurious jumps in the receiver gain 14 being controlled by the AGC loop 10. To be able to track such step changes in received signal strength with conventional AGC, the AGC loop bandwidth ordinarily would have to be kept larger, but that approach generally results in increased noise through the AGC loop 10.